6.4 Making Sense of Perception

What this lecture covers

This lecture closes the unit on perception by asking how, if we only ever have direct access to our own ideas or sense data, we can be entitled to believe there are mind-independent physical objects behind them. It works through Locke's indirect realism, a common misreading of it as trapping us behind a 'veil of ideas', and two rival twentieth-century responses: phenomenalism and the direct realism of Austin and Strawson.

By the end, you should be able to distinguish indirect from direct realism, explain the 'unacceptable interpretation' of ideas that Millican rejects, and describe how Hume's problem about the perception-object link resurfaces in each proposed solution.

Key ideas

• Indirect realism: Locke's view that we are directly aware of ideas in the mind and infer that a material object causes them, which raises the veil of perception problem: how do we know there is really an object behind the veil?
• The unacceptable interpretation: the mistaken picture of perception as a homunculus inside the head looking at a mental screen; this explains nothing, since it just relocates the problem of seeing a tree to the problem of seeing an image of a tree.
• Sense data: the twentieth-century term for how objects appear to us; better understood as awareness that counts as perception of an external object, not as an intermediate object that is itself perceived.
• Hume's challenge: because we are only ever acquainted with our own perceptions, experience can never establish a causal connection between perceptions and the external objects supposed to cause them.
• Phenomenalism: the view that statements about physical objects can be translated into statements about what sense data we would experience under certain conditions, aiming to sidestep the question of objects existing behind perceptions.
• Direct realism: the Austin/Strawson view that we perceive objects themselves, not ideas of them; it defeats the unacceptable interpretation but does not by itself answer the sceptic about whether a tree is really there.
• Ideas as intentional objects: a revised, less picture-like way of understanding a Lockean idea, as capturing what it is like to perceive an object rather than as a little image of it.
• Justifying belief in objects: the lecture's positive proposal is that we are entitled to believe in external objects because attributing causal, broadly structural properties to them lets us predict our future perceptions reliably.

Walkthrough

Indirect realism and the veil of perception (0:10)

Locke's indirect realism holds that perceiving a tree means being directly aware of an idea in the mind, caused by a material object. This raises the question of how we can know there is really an object behind the 'veil' of our ideas, and whether this traps us within our own mental states.

Rejecting the unacceptable interpretation (1:12)

Millican dismantles a parody of indirect realism in which a homunculus inside the head looks at a mental screen showing an image of the tree. This explains nothing, since it just replaces the problem of perceiving the tree with the problem of perceiving the image. The puzzle of why we do not see things upside down, despite the inverted retinal image, dissolves once this picture is dropped, and the same caution applies to talk of 'sense data': a sense datum should be understood as the way an object appears, not a further thing that must itself be perceived.

Hume's problem and phenomenalism (3:12)

Hume argues that since we are only ever aware of our own perceptions, we can never experience the causal link between perceptions and the external objects supposed to produce them, so the connection has no foundation in reasoning. Phenomenalism responds by treating statements about physical objects as statements about the sense data we would perceive under certain conditions, trying to sidestep the question of what lies behind perception. The lecture notes that phenomenalism still faces a version of the problem of induction, since it must assume that predicted future perceptions will actually occur, and that this approach fell out of favor after being popular in the mid-twentieth century.

Direct realism and its limits (8:20)

Austin and Strawson's direct realism insists we perceive objects themselves, not ideas of them, which correctly rules out the unacceptable interpretation: when you look at a tree, it is the tree you see, not an image of one. But this does not defeat the sceptic, who can still ask how you know your experience is actually caused by a tree rather than, say, being a brain in a vat. Direct realism, the lecture argues, gives a verbal solution rather than a genuine one.

Rehabilitating indirect realism (11:25)

The lecture proposes returning to a Lockean position, but reinterpreting an 'idea' not as a mental image, but as an intentional object: what it is like to perceive a tree, distinguishable in principle from the tree's actual existence. This keeps the logical possibility of hallucination or being a brain in a vat open without collapsing into the unacceptable interpretation.

Justifying belief in external objects (13:28)

To answer Hume's challenge, the lecture argues we are entitled to believe in external objects because attributing causal properties to them, transmitted through mechanisms like light and sound waves, lets us successfully predict future perceptions. We can give up the earlier assumption that these real properties must resemble our ideas, and accept explanations in terms of things like charge and spin, while still treating predictive success as good evidence that something like our conception of objects really exists.

Before you watch

• Watch the earlier lectures in this unit on primary and secondary qualities, since this lecture assumes familiarity with Locke's account of ideas and qualities.
• Recall Hume's argument about causal connections and experience, discussed elsewhere in the course.

Check your understanding

1. What is the veil of perception problem, and why does indirect realism raise it?
2. Explain the 'unacceptable interpretation' of perception and why Millican rejects it.
3. How does phenomenalism try to answer Hume's challenge, and what problem does it still face?
4. Why does direct realism fail to defeat the sceptic, even though it avoids the unacceptable interpretation?
5. How does the lecture propose we justify belief in external objects, given Hume's challenge?